**** BEGIN LOGGING AT Fri Aug 15 03:00:00 2014
Aug 15 03:13:07 <joran_> is there a good tutorial to create a recipe from qmake?
Aug 15 03:13:15 <joran_> for qt gui
Aug 15 03:42:12 <mranostay> hi rburton
Aug 15 03:54:23 <joran_> found this https://community.freescale.com/thread/309578 to answer my own question
Aug 15 03:57:18 <lpapp> joran_: there is a recipe for that, already.
Aug 15 04:12:43 <joran_> yeah that link i posted seems to make it pretty straightforward
Aug 15 04:13:18 <joran_> i just get nervous everytime i have to do something for the first time on yocto :P
Aug 15 04:13:26 <joran_> but im slowly getting the hang of it
Aug 15 04:20:26 <joran_> is this anything to be worried about? https://gist.github.com/anonymous/594a1ffc65e863a4e3cc
Aug 15 07:29:28 <lpapp> good morning
Aug 15 08:28:53 <lpapp> fray: I start thinking that you have a point with debug image for debugging.
Aug 15 08:30:04 <lpapp> I should probably apply that technique.
Aug 15 08:46:58 <Saur> Say I have a package A that I have added to IMAGE_INSTALL and it in turn rdepends on package B, then if I build my image all is fine and both packages are built and end up in the image. Then I update the version of package B and re-run bibtake. Now it rebuilds B, but the image is not updated. Is this a bug or am I missing some subtle detail in how dependencies work?
Aug 15 08:52:39 <bluelightning> morning all
Aug 15 08:57:46 <lpapp> bluelightning: the _exclude/remove option was introduced post-dylan for bitbake variables, yeah?
Aug 15 08:58:32 <JaMa> dora
Aug 15 08:58:43 <bluelightning> so yes
Aug 15 08:59:33 <bluelightning> Saur: that sounds broken to me...
Aug 15 08:59:42 <bluelightning> Saur: did do_rootfs run or not?
Aug 15 08:59:56 <Saur> bluelightning: Nope, do_rootfs did not run.
Aug 15 09:00:23 <bluelightning> that suggests that somehow the signature did not change
Aug 15 09:00:29 <bluelightning> which shouldn't be possible...
Aug 15 09:00:57 <lpapp> thanks
Aug 15 09:01:12 <bluelightning> Saur: is that behaviour reproducible?
Aug 15 09:01:19 <Saur> bluelightning: Yes
Aug 15 09:01:29 <bluelightning> Saur: ok, could you please file a bug?
Aug 15 09:02:20 <Saur> bluelightning: Well, it is reproducible for me at least. Haven't checked if any of my colleagues can reproduce it. Will do that first...
Aug 15 09:03:10 <bluelightning> Saur: ok, but even if not it would be worth trying to figure out why it's behaving like that on your system
Aug 15 09:03:30 <Saur> bluelightning: Oh, definitely. :)
Aug 15 09:04:45 <Saur> bluelightning: Any tips or tricks on how to dive into the dependency mechanism of bitbake to try to find out why the image doesn't think it needs to be rebuilt?
Aug 15 09:05:12 <bluelightning> Saur: bitbake-diffsigs would be my tool of choice for that
Aug 15 09:05:25 <bluelightning> we also now have bitbake -S printdiff
Aug 15 09:05:55 <Saur> bitbake-diffsigs I've used before. What does bitbake -S printdiff do?
Aug 15 09:16:28 <bluelightning> Saur: AFAIK it looks for the closest signature match in the sstate cache and then does a bitbake-diffsigs to show what the difference is
Aug 15 09:16:45 <Saur> bluelightning: Ok, sounds good.
Aug 15 09:16:46 <bluelightning> i.e. tries to answer "why wasn't sstate reused?"
Aug 15 11:04:26 <lpapp> JaMa: seems it has not been enough to use this to get the latest source:
Aug 15 11:04:26 <lpapp> SRCREV = "${AUTOREV}"
Aug 15 11:04:27 <lpapp> PV = "${SRCPV}"
Aug 15 11:04:45 <lpapp> JaMa: bitbake foo-git was doing the do_fetch stage, but it does not seem contain the latest commit.
Aug 15 11:04:51 <lpapp> I waited a couple of minutes after the push.
Aug 15 11:05:34 <lpapp> ah, sorry, my bad, I have not checked out the newly created AUTOINC folder.
Aug 15 11:06:10 <lpapp> is there an option to clean up the old stuff automatically?
Aug 15 11:06:21 <lpapp> and I do not mean rm_work here cause I would like to keep the latest to hack on.
Aug 15 11:07:22 <rburton> rm_old_work
Aug 15 11:08:52 <lpapp> hidden gem, yeah? :)
Aug 15 11:09:20 <lpapp> I cannot find anything like that in meta/ though.
Aug 15 11:09:25 <lpapp> not even scripts or bitbake
Aug 15 11:09:41 <rburton> its not in oe-core
Aug 15 11:12:28 <lpapp> why not ?
Aug 15 11:12:38 <lpapp> it sounds useful for removing old git stuff while preserving the newest.
Aug 15 11:13:46 <rburton> because nobody has sent it
Aug 15 11:14:11 <rburton> i'd like to see rm_work merge the abilitiies that rm_old_work has
Aug 15 11:15:15 <stiandre> anyone having experience adding support for freescale flexcan (driver and device) on imx6 (I am using wandboard-solo)?
Aug 15 11:19:39 <lpapp> rburton: in what form?
Aug 15 11:27:39 <JaMa> lpapp: doing both cleanups in single .bbclass maybe with variable to disable one of them if requested
Aug 15 14:46:15 <joran_> ok another dumb question ....
Aug 15 14:46:38 <joran_> ive been following these instructions for setting up yoctoqt
Aug 15 14:46:39 <joran_> https://community.freescale.com/thread/309578
Aug 15 14:47:11 <joran_> but I dont have any file like /opt/poky/1.4.1/environment-setup-armv7a-vfp-neon-poky-linux-gnueabi
Aug 15 14:47:35 <joran_> (I am using poky/1.6.1 but I dont have any environment-setup*
Aug 15 14:48:59 <rburton> those instructions are specific to the freescale bsp
Aug 15 14:49:34 <rburton> oh no, actually that's the ADT
Aug 15 14:49:44 <joran_> so if im using gumstix what does that mean for me?
Aug 15 14:50:02 <rburton> those instructions are terrible
Aug 15 14:50:16 <rburton> did you follow the first few steps?
Aug 15 14:50:23 <rburton> thats were you get environment-setup-* from
Aug 15 14:51:32 <joran_> yeah it built but I only have
Aug 15 14:51:51 <rburton> did you then install it?
Aug 15 14:52:14 <rburton> what that's doing is building a standalone toolchain and then telling qtcreator to use it
Aug 15 14:52:37 <joran_> ahh I lied i guess whatever I had running last night made the environfile
Aug 15 14:52:41 <joran_> im an idiot
Aug 15 14:53:37 <joran_> I swear it wasnt there when I went to bed ... but i was bitbaking qt4-embeded which must have spawned that
Aug 15 14:53:57 <joran_> but meta-toolchain-qt didnt I suppose
Aug 15 16:39:27 <hbruce> khem: ran clean command as you suggested but seeing same error. I'm running with -j8 (autodetect)
Aug 15 16:41:28 <lpapp> YAB became a bit floody ...
Aug 15 16:42:28 <rburton> lpapp: master-next just finished
Aug 15 16:43:15 <rburton> but as SWAT this week i'm glad its mostly success right now
Aug 15 16:44:47 <lpapp> :-)
Aug 15 17:34:41 <geckos> I saw this comment on some template recipe for kernel module, `# The inherit of module.bbclass will automatically name module packages with "kernel-module-" prefix as required by the oe-core build environment.', Should I rename my module to kernel-module something or this should work out of box?
Aug 15 17:37:26 <kergoth> it says it will automatically name it that
Aug 15 17:37:31 <kergoth> i take that to mean it will automatically name it that
Aug 15 18:09:34 <WarheadsSE> Anyone ever see a repo sync barf on a tag .. despite it being there? I've tried vXXX-thing-release refs/tags/vXXX-thing-release ..
Aug 15 18:10:02 <WarheadsSE> Keep ending up with error: in `sync`: revision vXXX-thing-release in user/repo not found
Aug 15 18:10:53 <geckos> How can I tell yocto to install kernel headers on sysroot?
Aug 15 18:11:09 <Crofton|work> anyone know why perf has symbols for libc and kernel, but not a user library?
Aug 15 18:11:30 <geckos> I put a header-y += myheader.h on properly Kbuild, and run bitbake -fc install linux, but my header isn't installed
Aug 15 18:12:25 <kergoth> ;the install task doesn't write to the sysroot :P
Aug 15 18:12:29 <kergoth> run populate_sysroot
Aug 15 18:57:22 <volker> does anyone know an easy way to identify which kind of hardware watchdog a board has? The Inel NUC specification does not list it.
Aug 15 18:57:35 <volker> compiling all hardware watchdogs into the kernel to see which one shows up?
Aug 15 20:41:04 <neodypsis> Hello. What may be the cause for "/sbin/init" to be missing from the image? It's causing  a kernel panic.
Aug 15 20:43:46 <theguy> Hey guys, I'm back with the same problem
Aug 15 20:45:35 <theguy> My system is still creating /dev/sda on boot even when no drive is inserted
Aug 15 21:01:48 <volker> neodypsis: how do you build it?
Aug 15 21:18:51 <neodypsis> volker: using bitbake my-yimage
Aug 15 21:19:26 <neodypsis> then I used the .hddimg
Aug 15 21:28:19 <volker> ok, I never used the hddimg. did you remove any packages? I never had problem with the tiny example and then extending it with the packages I need.
Aug 15 21:28:28 <volker> which version do you use?
Aug 15 21:29:15 <neodypsis> yes, I've been trying different configurations adding/removing packages.  I'm using Poky 1.6
Aug 15 21:34:47 <neodypsis> should I delete TMP and try building again?
Aug 15 21:35:29 <Guest24562> hi
Aug 15 21:36:12 <Guest24562> does anyone know why 40+ recipes get built when I try to build a single-source really simple and dumb recipe?
Aug 15 21:36:31 <Guest24562> I have my recipe called skeleton_1_0.bb
Aug 15 21:36:38 <Guest24562> put that into a new layer
Aug 15 21:36:51 <Guest24562> added layer to local.conf and removed other layers
Aug 15 21:36:57 <Guest24562> but when i do bitbake -g skeleton
Aug 15 21:37:00 <Guest24562> I get this list
Aug 15 21:37:08 <Guest24562> 1 xz-native   2 m4-native   3 skeleton   4 prelink-native   5 bison-native   6 readline-native   7 eglibc   8 gnu-config-native   9 pseudo-native  10 binutils-native  11 beecrypt-native  12 attr-native  13 elfutils-native  14 eglibc-initial  15 pkgconfig-native  16 gcc-cross-initial  17 libtool-native  18 gcc-runtime  19 libgcc  20 gettext-minimal-native  21 linux-libc-headers  22 quilt-native  23 zlib-native  24 ncurses-native  2
Aug 15 21:37:42 <Guest24562> there must be some default stuff set here
Aug 15 21:38:24 <Guest24562> any points will be appreciated
Aug 15 21:42:41 <Guest24562> hi, any pointers to beginners?
Aug 15 21:43:00 <Guest24562> I've looked through the docs, but the ramp still seems quite steep
Aug 15 21:44:32 <JaMa> Guest24562: INHIBIT_DEFAULT_DEPS
Aug 15 21:44:59 <JaMa> and BASEDEPENDS
Aug 15 21:45:04 <JaMa> are your pointers
Aug 15 21:48:33 <Guest24562> thx JaMa! going to give it a try
Aug 15 21:52:35 <volker> neodypsis: deleting TMP should not matter, yocto rebuilds automatically if needed
Aug 15 21:52:57 <volker> neodypsis: I would try first poky-tiny and see if you have the issue (I never had it with 1.6) and then go one step after another to see what broke it.
Aug 15 21:53:07 <volker> neodypsis: unless it is a bug...
Aug 15 21:55:25 <neodypsis> volker: ok, will try that, thx
Aug 15 22:04:02 <Guest24562> so I was able to reduce the size of tasks
Aug 15 22:04:08 <Guest24562> to 33
Aug 15 22:04:29 <Guest24562> this is by setting INHIBIT_DEFAULT_DEPS = "1", DEPENDS = "", BASEDEPENDS = ""
Aug 15 22:04:34 <Guest24562> and RDEPENDS = ""
Aug 15 22:04:55 <Guest24562> but I still get packages like binutils and ncurses by default...
Aug 15 22:06:15 <Guest24562> I believe setting DEPENDS and RDEPENDS to emptry sting would remove dependencies for this recipe
Aug 15 22:06:20 <Guest24562> but apparently not...?
Aug 15 22:11:11 <kergoth> just because your recipe doesn't deepnd on anything doesn't mean we don't need to build anything
Aug 15 22:11:20 <kergoth> if your recipe is packaged, then we need to build packaging tools and all their deps
Aug 15 22:11:32 <kergoth> (and every recipe is packaged if a packaging class is inherited)
Aug 15 22:12:00 <kergoth> you can use bitbake -g -u depexp recipename to examine the dependency graph
Aug 15 22:13:48 <Guest24562> I see. I guess I'm not seeing the whole picture here yet.
Aug 15 22:14:17 <Guest24562> but first of all I need to resolve this thing with "gobject" not being found ... I'm using CentOS 6.5
Aug 15 22:14:22 <Guest24562> brb
**** ENDING LOGGING AT Sat Aug 16 02:59:58 2014